INTRODUCTION:Post-operative cognitive dysfunction (POCD) is frequent in patients older than 60 years undergoing major non-cardiac surgery, and increases both morbidity and mortality. Anesthetic drugs may exert neurotoxic effects and contribute to the genesis of POCD. The hypothesis of the POCD-ELA trial was that closed-loop target-controlled infusion of propofol and remifentanil could reduce the occurrence of POCD by decreasing the risk of excessive depth of anesthesia and the dose of anesthetic drugs. METHODS AND ANALYSIS:We designed a single-center, single-blind, randomized, controlled, parallel trial and aim to include 204 patients aged >60 years undergoing elective major non-cardiac surgery. Patients will be randomized to receive closed-loop versus manual target-controlled infusion of propofol and remifentanil guided by bispectral index monitoring. Cognitive assessment will be performed the day before surgery (baseline) and within 72 hours after surgery, using a battery of validated neuropsychological tests. The primary outcome is the incidence of POCD within 72 hours after surgery. POCD is defined as a Z-score value > 1.96 for at least 2 different tests or a Z-score composite value >1.96. The calculation of the Z-score is based on data from an age-matched control population who did not undergo surgery or general anesthesia. ETHICS AND DISSEMINATION:This study was approved by the Ethics Committee (Comité de Protection des Personnes Est-II) and authorized by the French Health Products Agency (Agence Nationale de Sécurité des Médicaments, Saint-Denis, France). The University Hospital of Besancon is the trial sponsor and the holder of all data and publication rights. Results of the study will be submitted for publication in a peer-review international medical journal and for presentation in abstract (oral or poster) in international peer-reviewed congresses. REGISTRATION:The trial is registered with ClinicalTrials.gov (Identifier: NCT02841423, principal investigator: Prof Emmanuel Samain, date of registration: July 22, 2016). Last amendment of protocol: version 8.0 April 2018.